END序列下半参数回归模型估计的相合性

摘要

We studied the semiparametric regression model yi = xiβ + g ( ti ) + σiεi, i =1,2,…, n, integrating the methods of weighted function and least squares,we established the least squares estimator and weighted least squares estimator of the model′s parameter β and the unknown function g. By using the truncating method and the moment inequalities for END sequences, we got the p⁃th (p>1) mean consistency.%研究误差为END序列的半参数回归模型yi ＝xiβ＋g（ti）＋σiεi（i＝1，2，…，n）。应用加权估计与最小二乘估计方法，建立未知参数β和未知函数g的最小二乘估计与加权最小二乘估计的估计量。利用END序列的Rosenthal不等式以及截尾的方法证明p（p＞1）阶矩的相合性。